Abstract: An apparatus, method and computer readable medium for a voice-controlled camera with artificial intelligence (AI) for precise focusing. The method includes receiving, by the camera, natural language instructions from a user for focusing the camera to achieve a desired photograph. The natural language instructions are processed using natural language processing techniques to enable the camera to understand the instructions. A preview image of a user desired scene is captured by the camera. Artificial Intelligence (AI) is applied to the preview image to obtain context and to detect objects within the preview image. A depth map of the preview image is generated to obtain distances from the detected objects in the preview image to the camera. It is determined whether the detected objects in the image match the natural language instructions from the user.

Abstract: A vehicle relocator may include one or more processors, which are configured to receive first sensor data representing a current or predicted weather condition; receive second sensor data representing at least position or a vicinity of a vehicle; determine from at least the first sensor data and the second sensor data a risk of damage to the vehicle; and if the determined risk is outside of a target range, send an instruction for the vehicle to travel to an alternate location.

Abstract: An apparatus, method and computer readable medium for a voice-controlled camera with artificial intelligence (AI) for precise focusing. The method includes receiving, by the camera, natural language instructions from a user for focusing the camera to achieve a desired photograph. The natural language instructions are processed using natural language processing techniques to enable the camera to understand the instructions. A preview image of a user desired scene is captured by the camera. Artificial Intelligence (AI) is applied to the preview image to obtain context and to detect objects within the preview image. A depth map of the preview image is generated to obtain distances from the detected objects in the preview image to the camera. It is determined whether the detected objects in the image match the natural language instructions from the user.

Abstract: One embodiment of a virtual reality apparatus comprises: a graphics processing engine comprising a plurality of graphics processing stages, the graphics processing engine to render a plurality of image frames for left and right displays of a head mounted display (HMD); and foveation control hardware logic to independently control two or more of the plurality of graphics processing stages based on feedback received from an eye tracking module of the HMD, the feedback indicating a foveated region selected based on a current or anticipated direction of a user's gaze, the foveation control hardware logic to cause the two or more of the graphics processing stages to process the foveated region differently than other regions of the image frames.

Abstract: According to various aspects, a collision avoidance method may include: receiving depth information of one or more depth imaging sensors of an unmanned aerial vehicle; determining from the depth information a first obstacle located within a first distance range and movement information associated with the first obstacle; determining from the depth information a second obstacle located within a second distance range and movement information associated with the second obstacle, the second distance range is distinct from the first distance range, determining a virtual force vector based on the determined movement information, and controlling flight of the unmanned aerial vehicle based on the virtual force vector to avoid a collision of the unmanned aerial vehicle with the first obstacle and the second obstacle.

Abstract: Apparatus and method for correcting image regions following upsampling or frame interpolation. For example, one embodiment of an apparatus comprises a machine-learning engine to evaluate at least a first image in a sequence of images generated by a real-time interactive application, the machine learning engine to responsively use previously learned data to generate an upsampled or interpolated image comprising a plurality of pixel patches. In one embodiment, each pixel patch is associated with a confidence value reflecting how accurately the pixel patch was generated by the machine learning engine. A selective ray tracing engine identifies a first pixel patch to be corrected based a first confidence value corresponding to the first pixel patch being lower than a threshold and performs ray tracing operations on a first portion of the first image to generate a corrected first pixel patch.

Abstract: An infrastructure state prediction device includes a processor configured to receive sensor information from a sensor at an infrastructure element, wherein the sensor information includes an observation of a traffic object at the infrastructure element; and determine, based on the sensor information and an infrastructure state model comprising information indicative of a state timing for the infrastructure element, an updated state timing for the infrastructure element; and a transmitter configured to transmit the updated state timing to the infrastructure element.

Abstract: The disclosed embodiments generally relate to methods, systems and apparatuses to provide ad hoc digital signage for public or private display. In certain embodiments, the disclosure provides dynamically formed digital signage. In one application, one or more drones are used to project the desired signage. In another application one or more drones are used to form a background to receive the projected image. In still another application, sensors are used to detect audience movement, line of sight or engagement level. The sensor information is then used to arrange the projecting drones or the surface-image drones to further signage presentation.

Abstract: An apparatus to facilitate deep learning based selection of samples for adaptive supersampling is disclosed. The apparatus includes one or more processing elements to: receive training data comprising input tiles and corresponding supersampling values for the input tiles, wherein each input tile comprises a plurality of pixels, and train, based on the training data, a machine learning model to identify a level of supersampling for a rendered tile of pixels.

Abstract: Systems, apparatuses, and methods may provide for technology to dynamically control a display in response to ocular characteristic measurements of at least one eye of a user.

Abstract: The present disclosure is directed at pairing a host electronic device with a peripheral electronic device using visual recognition and deep learning techniques. In particular, the host device may receive an indication of a peripheral device via a camera or as a result of searching for the peripheral device (e.g., due startup of a related application or periodic scanning). The host device may also receive an image of the peripheral device (e.g., captured via the camera, and determine a visual distance to the peripheral device based on the image. The host device may also determine a signal strength of the peripheral device, and determine a signal distance to the peripheral device based on the signal strength. The host device may pair with the peripheral device if the visual distance and the signal distance are approximately equal.

Abstract: A system for unmanned aerial vehicle alignment including: an image sensor, configured to obtain an image of unmanned aerial vehicles and provide to a processor image data corresponding to the obtained image, the processor, configured to determine from the image data image positions of the unmanned aerial vehicles, determine an average position of the unmanned aerial vehicles relative to a first axis based on the image positions, determine an average line that extends along a second axis through the average position, wherein the first and second axes are perpendicular to each other, determine a target position of one of the unmanned aerial vehicles based on a relationship between its respective image position and a target alignment, and determine an adjustment instruction to direct said one of the unmanned aerial vehicles toward the target position, and provide the adjustment instruction to said one of the unmanned aerial vehicles.

Abstract: Herein is disclosed an infrastructure state prediction device comprising a memory device onto which an infrastructure state model corresponding to a location and a state timing of a plurality of infrastructure elements is stored; one or more processors, communicatively coupled to the memory device and configured to receive measurement information representing a location and an observed state of a first infrastructure element; determine a predicted state of a second infrastructure element based on a state timing corresponding to a second infrastructure element; determine a movement information based on the observed state of the first infrastructure element, the predicted state of the second infrastructure element and a relation between the location of the first infrastructure element and a location of the second infrastructure element; and generate a message comprising the movement information.

Abstract: Methods and apparatus for cloud gaming Graphics Processing Unit (GPU) with integrated Network Interface Controller (NIC) and shared frame buffer access. The GPU include one or more frame buffers that provide shared access to an integrated encoder/decoder. The GPU further includes an integrated NIC coupled to the integrated encoder/decoder and one or more video outputs coupled to the one or more frame buffers. The GPU is configured to process outbound and inbound game image content that is encoded and decoded using a video codec or using a game tile encoder and decoder. Video frames buffered in the frame buffer(s) are encoded by the integrated encoder and forwarded directly to the NIC to be packetized and streamed using a media streaming protocol. Inbound streamed media content is depacketized by the NIC and decoded by the integrated decoder, which writes the decoded content to a frame buffer to regenerate.

Abstract: Herein is disclosed an unmanned aerial vehicle comprising a memory, configured to store a projection image; an image sensor, configured detect image data of a projection surface within a vicinity of the unmanned aerial vehicle; one or more processors, configured to determine a depth information for a plurality of points in the detected image data; generate a transformed projection image from a projection image by modifying the projection image to compensate for unevenesses in the projection surface according to the determined depth information; and send the transformed projection image to an image projector; and an image projector, configured to project the transformed projection image onto the projection surface.

Abstract: According to various aspects, a vehicle may include: an external structure; an assembly configured to allow a movement of at least a part of the assembly relative to the external structure at least from a first position to a second position, wherein the assembly protrudes from the external structure with a first distance in the first position and with a second distance less than the first distance in the second position; one or more sensors configured to receive obstacle information associated with one or more obstacles in a vicinity of the vehicle; one or more processors configured to determine a collision threat to the assembly based on the obstacle information, and trigger the movement of at least the part of the assembly from the first position into the second position in the case that the collision threat is determined.

Abstract: Methods and apparatus are disclosed for using long exposure video for virtual reality headsets. An example apparatus includes a video camera to capture first, real time video of a scene using a first field of view while operating using a first exposure time, the first exposure time corresponding to a normal exposure time of a human eye; a transformer to transform the first, real time video into a second video corresponding to a second exposure time, the second exposure time being at least twice as long as the first exposure time; and a display to display the second video at the second exposure time.

Abstract: System and techniques for vehicle-based measurement of signal object integrity are described herein. Sensor data of an environment of the vehicle is obtained. A bound for the sensor data for the sensor data is also obtained. Here, the bound corresponds to a signal object (e.g., sign, light, road marking, etc.). A classifier for the signal object is invoked on the sensor data based on the bound. If the classifier produces a result that indicates a problem with the signal object, a representation of the result is communicated.

Abstract: An embodiment of a semiconductor package apparatus may include technology to analyze an electronic image to determine indirect information including one or more of shadow information and reflection information, and provide the indirect information to a vehicle guidance system. Other embodiments are disclosed and claimed.

Abstract: Various systems and methods for multimodal trip planning are described herein, including determining candidate travel paths for a trip request, determining a score for each of the determined travel candidate paths, and determining adjusted scores for each of the candidate travel paths using an indication of physiological ability of the user.